基于混合高斯过程的多模型热力参数测量软仪表

THERMAL PARAMETER SOFT SENSOR BASED ON THE MIXTURE OF GAUSSIAN PROCESSES

为了解决工业过程中参数失效和优化运行中参数计算的问题,提出了一种新型软仪表,它是基于混合高斯模型,利用EM算法实现对混合模型中的参数估计。混合模型的使用既有利于降低单一高斯模型的计算负担,又能有效反映工业过程中的工况变化,判断出与特定工况相关程度最大的过程知识,利用它们建立与特定工况对应的局部模型,并将它们合并组成具有多模型结构的全局模型。作为示例,建立了测量火电厂烟气含氧量的软仪表。仿真结果表明,文中提出的方法能有效地实现工业过程参数的软测量,具有较大的实用价值。

In order to solve the problem of the failure of measure parameters and online optimal running in industrial processes, a novel soft sensor is investigated which is based on mixtures of Gaussian processes(GP) with expectation maximization(EM) algorithm employed for parameter estimation of mixture of models. The mixture model can alleviate computational complexity of GP and also accord with changes of operating condition in industrial processes, i.e., it would certainly be able to examine what types of process-knowledge would be most relevant for specific operating points of the process by local models and then combine them into a global one. Finally, soft sensor for measuring O2 content in flue gas is demonstrated by a real-world example. Simulations show that the proposed method can effectively implement soft sensor for industrial process parameters.